openarm
A fully open-source humanoid arm for physical AI research and deployment in contact-rich environments.
- 3.7k
- MDX
- Apache-2.0
- 6 d ago
A fully open-source humanoid arm for physical AI research and deployment in contact-rich environments.
Humanoid-Gym: Reinforcement Learning for Humanoid Robot with Zero-Shot Sim2Real Transfer https://arxiv.org/abs/2404.05695
MultiModalWBC is a fully open-source, IsaacLab-based framework for multi-modal whole-body control, designed for motion imitation, motion tracking, and task-conditioned control in legged robots. The framework unifies robot proprioceptive states and multi-modal human motion conditions into a consistent interface
🤖 Full featured 21 DOF 3D Printed Humanoid Robot based on ATmega328P chip
Open-source Otto robot + XiaoZhi AI technical documentation and sharing
Deployment-focused fork of OpenPI for LimX TRON2 manipulation — pi0.5 policy serving, task fine-tuning, TRON2 transforms, and real-robot client examples.
Official home of the LimX Dynamics open-source ecosystem — humanoid robots, legged robots, locomotion & manipulation, and embodied AI.
Python RL deployment for LimX humanoid — ONNX runtime, MuJoCo sim & real-robot
URDF/xacro, MuJoCo XML, mesh & USD models for LimX TRON2A & TRON2B — 16 variants (dual-arm, bipedal, wheeled-leg), ROS 1/2 & simulation ready.
Official code for "Vision as Proprioception: Estimating Humanoid Robot Pose from Video"
RL training pipeline for LimX TRON2A humanoid — Isaac Lab + PPO (rsl_rl); sole-foot, wheel-foot and whole-body dual-arm DASF variants.
URDF, MuJoCo MJCF & USD models for the LimX Oli humanoid (HU_D04, plus legacy HU_D03). Includes gripper and dexterous-hand variants. ROS 1/2, Isaac Sim & MuJoCo ready.
ROS 2 Humble RL deployment for LimX humanoid — ONNX inference & motion control
MuJoCo simulation for LimX humanoid — controller testing & sim2real policy validation
Papers, code, competition results and partner deployments built on LimX robots (TRON1, TRON2, Oli, Luna, P1, W1). Browse the gallery or read the full list.
ROS-free C++ RL deployment for LimX humanoid — ONNX inference & motion control
MuJoCo simulator for the LimX TRON2 family — bridges SDK RobotCmd/RobotState to physics for controller testing and Sim-to-Real verification. Covers SF, WF, DA, DACH and DASF as TRON2A and TRON2B.
Isaac Lab RL training stack for TRON2A with 6-DoF arm and gripper — PPO-based locomotion policy training supporting SF/WF (foot variants) and SFYG/WFYG (arm+gripper locked variants). Built for the ATEC challenge.
Isaac Lab RL training for LimX Oli humanoid — Isaac Sim 4.5, sim-to-real locomotion
LimX Dynamics fork of GMR (General Motion Retargeting): retarget human motion (SMPL-X/BVH/Xsens/OptiTrack) into LimX humanoids (OLI EDU, Luna) and 17+ other robots in real time on CPU.
Insights into the embodied AI open-source ecosystem by LimX Dynamics, built from public GitHub data. 具身智能开源生态观察:月报与行业趋势。
RL deployment/inference for LimX TRON2A humanoid — sole-foot, wheel-foot and dual-arm DASF ONNX policies via ONNX Runtime, on MuJoCo sim and real hardware.
Manipulation controller runtime for the LimX TRON2A dual-arm robot (DACH_TRON2A) — QP-based motion planning & PD control over the WebSocket SDK, with a local MuJoCo simulation workspace.
Gym-style runtime environment for LimX TRON2 manipulation — WebSocket robot transport, motion control & multi-camera observation.
具身智能与巡检机器人专利精选:按技术领域与公司双维度收录,附 Mermaid 全景脑图,README 由 yaml-readme 自动生成
A native desktop dashboard for controlling and visualizing the Unitree G1 humanoid robot.
ROS Noetic RL deployment for LimX humanoid — ONNX inference & motion control
URDF, MuJoCo MJCF & USD models for the LimX Luna humanoid robot — 27 actuated joints, with ankle/waist parallel linkages modeled in the MJCF and the closed-loop USD. ROS, MuJoCo & Isaac Sim ready.